PDF - ORIGINAL UNEDITED MANUSCRIPT

1

The functional analysis of the dihydroflavonol 4-reductase family of Camellia

sinensis: exploiting the key amino acids to reconstruct the reduction activity

Haixiang Ruan 1,2#

, Xingxing Shi1,3 #

, Liping Gao2#

, Arif Rashid2, Yan Li

2, Ting Lei

2, Xinlong

Dai3, Tao Xia

1 *, and Yunsheng Wang

1, 2*

1State Key Laboratory of Tea Plant Biology and Utilization, Anhui Agricultural University, Hefei,

Anhui 230036, China; 2School of Life Science, Anhui Agricultural University, Hefei, Anhui

230036, China； 3College of Tea Scinence, Guizhou University, Guiyang Guizhou 550025, China;

# These authors contributed equally to this work.

* Correspondence: [email protected]; [email protected]

Running title: Key amino acid controlling DFR reductase activity

Keywords: Camellia sinensis, dihydroflavonol 4-reductase, dihydroflavonols, enzymatic

characteristics, metabolic engineering, site-directed mutagenesis

© The Author(s) 2022. Published by Oxford University Press. All rights reserved. This is an Open

Access article distributed under the terms of the Creative Commons Attribution License

http://creativecommons.org/licenses/by/4.0/, which permits unrestricted reuse, distribution, and

reproduction in any medium, provided the original work is properly cited.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow

nloaded from https://academ

ic.oup.com/hr/advance-article/doi/10.1093/hr/uhac098/6572277 by guest on 26 June 2022

mailto:[email protected]

2

Haixiang Ruan - [email protected]

Xingxing Shi - [email protected]

Liping Gao - [email protected]

Arif Rashid - [email protected]

Yan Li - [email protected]

Ting Lei - [email protected]

Xinlong Dai - [email protected]

Tao Xia - [email protected]

Yunsheng Wang - [email protected]

Highlight: We reported the first dihydroflavonol 4-reductase (CsDFRs) class II/III genes in

plants. This not only contributes to the DFR gene evolution, but provides candidate DFR gene for

flavonoid metabolic engineering.

Abstract:

Anthocyanin and proanthocyanidins (PAs) are important components of flavonoids, secondary

metabolites in plants with a wide range of industrial and pharmaceutical applications. DFR

(dihydroflavonol 4-reductase) is a pivotal enzyme which plays an important role in the flavonoid

pathway. Here, four CsDFRs genes were isolated from Camellia sinensis and their overexpression

were analyzed in vitro and in vivo. Based on transcription and metabolic analyses, the CsDFRs

expression is closely consistent with catechins and PAs accumulation. Moreover, the enzyme

activity analyses revealed that the two recombinant proteins CsDFRa and CsDFRc exhibited DFR

activity, which converts dihydroflavonols into leucocyanidin in vitro, but not CsDFRb1 and

CsDFRb3. CsDFRa and CsDFRc overexpression in AtDFR mutants (tt3) revealed that CsDFRs

are involved in the biosynthesis of anthocyanins and PAs, as CsDFRa and CsDFRc not only

restored the petiole purple phenotype, but also the seed coat color. Site-directed mutagenesis

revealed that the two amino acid residues S117 and T123 of CsDFRa play a prominent role in

controlling DFR reductase activity. Enzymatic assays indicated that CsDFRa and CsDFRc

exhibited a higher affinity for DHQ and DHK, respectively, whereas CsDFRb1N120S

and

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow








3

CsDFRb1C126T

exhibited a higher affinity for DHM. Our findings present comprehensive

characterization of the DFRs from C. sinensis and shed light on their critical role in metabolic

engineering.

Introduction

Flavonoids, which include flavones, flavanols, flavonols, flavan-3-ols (catechins), anthocyanidins

and proanthocyanidins (PAs), are a class of naturally occurring secondary metabolites distributed

widely in different tissues of plants, such as leaves, flowers, stems, and fruits 1. Flavonoids are

also the most potent and health-promoting compounds. In epidemiological, clinical, and animal

studies, they fight against different diseases, such as cardiovascular disease, cancer, and other

disorders 2-4

. The metabolic and biosynthesis pathways of plant flavonoids have piqued interest in

the field of plant secondary metabolites due to their functionalities, compelling evidence of health

benefits, industrial applications, and incorporation into various food products.

The flavonoid biosynthesis pathway in numerous plant species has been extensively investigated

at the genetic, molecular and biochemical levels, with the identification of key functional genes

and transposable elements having been identified. Naringenin is an important precursor that can be

hydroxylated by flavonoid-3-hydroxylase (F3H) at the 3-position of the C ring or hydroxylated by

flavonoid-3'-hydroxylase (F3'H) or flavonoid-3'5'-hydroxylase (F3'5'H) to produce three types of

dihydroflavonols, including DHK (dihydrokaempferol), DHQ (dihydroquercetin) and DHM

(dihydromyricetin) 5. DFR (dihydroflavonol 4-reductase) is an enzyme that converts

dihydroflavonols to leucoanthocyanidins, which are precursors to anthocyanidins, flavan-3-ols,

and PAs (Figure 1A) 6.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



4

Figure 1. The flavonoid biosynthesis pathway and the end-product accumulation of flavonoids in C. sinensis. (A) The

flavonoid biosynthesis. DFR, dihydroflavonol 4-reductase; CHI, chalcone isomerase; ANS, anthocyanidin synthase; CHS,

chalcone synthase; F3’H, flavonoid 3’-hydroxylase; F3H, flavanone 3-hydroxylase; F3’5’H, flavonoid 3’5’-hydroxylase; FLS,

flavonol synthase; LAR, leucoanthocyanidin reductase; ANR, anthocyanidin reductase; (B) The proportion of various flavonoid

compounds. The proportion of flavonoid compounds were taken from Jiang 7.

DFR is a flavonoid biosynthesis key rate-limiting enzyme that belongs to the NADPH-dependent

epimerase/dehydratase family. The genes that code for the DFR enzyme have been characterized

and isolated from a variety of plant species, and their functions have been extensively studied 8-11

.

Several reports have shown that DFR is primarily involved in the color of the petals and seed coat.

The Arabidopsis DFR-deficient mutant (tt3), for example, has a complete lack of PAs

accumulation and the seed coat becomes transparent, and the DFR-deficient gerbera has a normal

phenotypic of white-colored petals 12,13

. Other DFR-like genes in plants, such as Oryza sativa’s

OsDFR2 and DRL1 from Arabidopsis, have been established to have virous biochemical and

physiological activities in plants, and are essential for pollen formation and male fertility.

Several studies have shown that dihydroflavonol 4-reductases from different species have distinct

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



5

substrate specificity, such as the DFRs from Cymbidium 14

and Petunia 15

, which did not

effectively reduce DHK, the precursor to orange pelargonidin-type anthocyanins. Previously, a 26

amino acid region in the middle of the sequence was identified as determining the substrate

specificity of DFR proteins 15

.

C. sinensis is a commercially significant plant that produces one of the most popular non-alcoholic

beverages on the planet. Besides its distinct flavor, tea is regarded as a health drink due to its

nutritional and therapeutic properties 16,17

. Tea has become one of the most prominent symbols of

Chinese culture. Tea health promoting properties have been mainly attributed in recent years to its

polyphenol content, particularly flavan-3-ols (catechins) and flavonols, which comprise 30% of

fresh leaf dry weight 7,18

. Catechins with gallic acid esters, such as epigallocatechin-3-gallate

(EGCG) as well as epigallocatechin (ECG), are the most abundant catechins in tea and have

gained the most attention 19

. The total amount of PAs in the root, particularly proanthocyanidins

dimers and trimers, however, significantly higher than in the leaves and stems (Figure 1B and

Table S1) 7. Overall, these findings reveal that the tea plant accumulates a high amount of

catechins, but the molecular mechanism underlying catechin metabolism and regulation in tea

leaves remains elusive.

Previous studies indicate that CsDFRa is a key gene in regulating catechin content in tea plants 20

.

Previously, five CsDFR genes have been identified using genome sequence and transcriptome

databases 21

. In this study, four CsDFR genes (CsDFRa、CsDFRb1、CsDFRb3、CsDFRc) were

identified and cloned from the bud/root of C. sinensis cv. Shuchazao, and the expression patterns

in different tissues were analyzed using qRT-PCR. A phylogenetic analysis was performed which

revealed that CsDFRa was clustered into class Ⅰ, and the other three proteins of C. sinensis were

clustered independently outside the other NADPH-dependent reductase branches (class II and

class III). The prokaryotic expression vector was constructed and the enzyme activity was

determined in vitro. Moreover, to elucidate the function of CsDFRs in vivo, CsDFRs were

overexpressed in the Arabidopsis tt3 mutant to observe phenotypic and metabolic changes. To

determine the region of the DFRa and DFRb1 enzyme responsible for the substrate specificity, we

introduced chimeric DFRs and determined the activity region. Furthermore, we confirm that the

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



6

activity of DFR can be modulated by the alteration of the amino acids in the region. This paper

reports the functional identification of DFR class II and class III genes in plants comprehensively

and systematically, and it is proposed that these findings will not only aid in the study of the DFR

gene family evolution of in shrub/vine plants, but will also provide new candidate DFR genes for

flavonoid metabolic engineering.

Results

Cloning and sequence analysis of CsDFRs

Based on the genome sequence and transcriptome information of C. sinensis, five putative

CsDFRs genes were screened and identified, including CsDFRa (GeneBank ID: KY615690),

CsDFRb1 (KY615691), CsDFRb2 (KY615692), CsDFRb3 (KY615693), and CsDFRc

(KY615694) 21

. To clone the CsDFR genes, cDNA libraries were constructed from the buds and

roots of cv. Shuchazao and screened using the gene-specific primers of CsDFRs from C. sinensis

as listed in Table S2. The other four genes, except CsDFRb2, have been successfully cloned and

sequenced (Figure S1A). The sequencing results showed that the ORF of CsDFRa, CsDFRb1,

CsDFRb3 and CsDFRc are 1044bp, 1035bp, 1023bp and 1074 bp, respectively, encoding 347, 344,

340 and 357 amino acids (Figure S1A). The characteristics of four structural genes were shown in

Figure S1B. The four DFR genes have similar genetic structures with 6 exons and 5 introns,

similar to DFR genes from other plants such as Arabidopsis and buckwheat 22,23

. To understand

further the function of the DFR genes, we predicated the major transcription factor binding motif

in the promoter region (Figure S1C), and MYB binding sites have been detected in the promoter

region of DFRa, DFRb1, and DFRc. However, no MYB binding site was detected in the promoter

region of DFRb3. In the promoter region of DFRb1, the MYB motifs MYBPLANT and

MYBPZM were not found.

Sequence alignment with several identified DFRs revealed that the N-terminus regions of the C.

sinensis DFR proteins comprised putative NADPH and substrate binding sites (Figure 2A).

Compared with other DFR proteins, the substrate binding regions of the CsDFRb3 and CsDFRc

sequences have 5 or 3 redundant residues, respectively. DFR proteins can be classified into three

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



7

types based on the conservation of the 134th residue: asparagine (Asn), aspartic acid (Asp) and

neither-Asn nor Asp 15

. CsDFRa, like most DFR proteins from other species, is an Asn-type

protein whereas the other three proteins from C. sinensis are neither Asn nor Asp-type (Figure 2).

A phylogenetic tree was constructed using the neighbor-joining method to further investigate the

homology of these four proteins with other known DFRs and NADPH-dependent reductases, such

as anthocyanidin reductase (ANR), leucoanthocyanidin reductase (LAR), and anther-specific

DFR-like enzymes. The result showed that CsDFRa was clustered into class I, with proteins from

dicotyledon, monocotyledon, and gymnosperm species. In class I, several proteins have been

found in previous literature (Figure 2B). CsDFRa was most similar to Actinidia chinensis

(identified score: 98.86%), Rhododendron simsii, and Vaccinium ashei (89.28%). Furthermore, the

other three proteins of C. sinensis were clustered independently outside the other

NADPH-dependent reductase branches (class II and III), and CsDFRb1 and b3 were slightly

closer to the core DFR branch. To our knowledge, the functions of genes in classes II and III have

not been confirmed in vivo or vitro, and most of the genes are from shrubs or vines, especially

from Actinidia chinensis, Vitis vinifera, and Manihot esculenta. Therefore, CsDFRb1, b3, and

CsDFRc might have special significance for studying the evolution of DFR gene family. The

function of these genes needs to be further elucidated.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



8

Figure 2. Sequence analysis of CsDFRs. (A) Multiple sequence alignment of DFRs from various plant species. The DFR

featured motifs, including the NADPH binding region and substrate binding site are boxed in red and blue, respectively. (B)

Phylogenetic tree of CsDFRs and other reductase different plant species. The accession numbers were obtained from publically

accessible database NCBI, and are presented in the figure.

Expression pattern of CsDFRs in tea plants

Quantitative RT-PCR was used to detect the expression patterns of CsDFR genes in tea plants. The

gene-specific primers were used for the amplification of the expression tendency of four genes in

distinct tissues (buds and young leaf, mature leaf, stem and root) (Figure 3A). For normalization,

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



9

the gene GAPDH was used as a reference control. Each gene was differentially expressed in

distinct tissues (Figure 3B), such as transcripts level of CsDFRa were detected in all tissues

examined, with the highest expression level in the bud, followed by the young leaf and stem, and a

low expression level was detected in the mature leaf and root. The expression pattern of CsDFRa

is consistent with the concentration of catechins in tea plants (Figure 1B). However, the

expressions of other genes, such as CsDFRb1, CsDFRb3, and CsDFRc, were highest in the root

and lowest in the mature leaf and stem, which is consistent with the concentration of PA and

indicating the involvement of these three genes in the PA biosynthesis in tea plant. Moreover,

based on RNA sequencing and qRT-PCR results, the expression level of CsDFRa（the mean Ct：

19.67，the RPKM：696.54 in the bud) was significantly superior than the other three genes

(Dataset S1).

Figure 3. Tempo-spatial expression pattern of CsDFR genes in C. sinensis using Quantitative RT-PCR analysis (A) the

figure shows different tissues collected from numerous plants for gene expression. (B) Transcripts levels CsDFRs. The data are

represented as a mean and ± standard deviation of three biological replicates (n = 3), and GAPDH were used as internal control

for the normalization of the Ct values. The alphabets (a, b, c, d, e and f) represents the significance level at P＜0.05.

Recombinant CsDFRs proteins catalytic activity in vitro

To analyze further the enzymatic properties of CsDFRa, CsDFRb1, CsDFRb3, and CsDFRc, a

series of SUMO vectors carrying different CsDFR genes were constructed and successfully

expressed in E. coli. The SDS-PAGE electrophoretogram showed that clear target bands could be

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



10

obtained after IPTG inducement of recombinant strains (Figure 4A).

We detect the product anthocyanin, splitted from the leucoanthocyanidin under high temperature

and acidic conditions, because the leucoanthocyanidin produced by DFR enzyme is highly

unstable and difficult to detect directly by using HPLC. The red products were observed in the

culture of the recombinant strain carrying the CsDFRa or CsDFRc gene, using different substrates,

including DHK, DHQ, or DHM. As illustrated in Figure 4B, when dihydroflavonols were used as

a substrate, no red products were observed, indicating that CsDFRb1 and b3 had little or no

reductive activity.

Figure 4. Analyses of catalytic reaction product with CsDFRs fusion proteins. (A) Schematic map of the recombinant

CsDFR construct and the profiles of expressed CsDFRs fusion protein. M: Protein molecular weight markers; 1: Before

induction; 2: The total proteins of induction by SUMO-CsDFRs; 3: Purified protein. (B) Analyze of the enzyme reaction with

SUMO-CsDFRs in vitro, and DHK、DHQ and DHM as the substrate.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



11

Functional confirmation of the CsDFRs in Arabidopsis

In A. thaliana, tt3 was the DFR genes mutant, which not only showed a deficiency in anthocyanin

synthesis but also exhibited an absence of the seed coat pigment. Therefore, the tt3 mutant is a

suitable model plant for determining whether CsDFR is associate with the biosynthesis of

anthocyanin and PAs.

The ORFs of three CsDFRs (including CsDFRa, CsDFRb1, and CsDFRc) were transformed into

tt3 mutant driven by 35S promoter, to elucidate whether CsDFRs have a functional tt3 ortholog

that can restore the above-mentioned deficient phenotypes of Arabidopsis tt3 mutant. After

overexpression of CsDFRs in Arabidopsis tt3 mutant, perceptible phenotypic vary were observed

in the CsDFRa and CsDFRc transgenic seedlings and/or seed coats relative to the tt3 mutant

(Figure 5A). The CsDFRa and CsDFRc not only restored the petiole purple phenotype but also

recovered the seed coats color of the tt3 mutant, as compared to the tt3 mutant and CsDFRb1

transgenic lines (Figure 5A).

DFR converts dihydroflavonols into leucoanthocyanins in the flavonoid biosynthesis pathway.

Following that, leucoanthocyanins were further converted into anthocyanins by ANS. These

aforementioned compounds mentioned will play a role in the synthesis of PAs in plants 24,25

.

Polyphenols were extracted from wild type (Ler), tt3 mutant, and CsDFRs transgenic lines to

detect the contents of anthocyanins and soluble PAs in the seedlings and seeds. CsDFRa and

CsDFRc transgenic Arabidopsis had significantly higher anthocyanins content in the seedlings

than CsDFRb1 and tt3 mutant lines (Figure 5B). Furthermore, DMACA staining revealed that

bluish compounds were formed in the seeds extract of the CsDFRa and CsDFRc transgenic lines,

indicating that the PAs were increased in the seeds compared to the control tt3 mutant, along with

wild type as a positive control (Figure 5B). These findings suggested that CsDFRa and CsDFRc

were involved in the biosynthesis of anthocyanin and PAs, and they are consistent with the

catalytic function of CsDFRa and CsDFRc, which can catalyze the dihydroflavonols into

leucoanthocyanidins, but CsDFRb1 cannot in vivo.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



javascript:;

12

Figure 5. Overexpression CsDFRs in Arabidopsis mutant tt3. (A) The phenotype of the Ler, tt3, CsDFRa/tt3, CsDFRb1/tt3

and CsDFRc/tt3 in seeding or seeds. (B) The absorbance values of the anthocyanin extracts and of bluish compounds from

DMACA reactive compounds at 530 nm and 640 nm, respectively. All the values are the means of three biological replicates, and

the different letters (A, B, C and a, b, c) indicated the significant level at P＜0.05.

Identification of a region that determines the activity deference between CsDFRa and

CsDFRb1

Although the amino acid sequences of the CsDFRb1 and CsDFRa are quite similar, the CsDFRa

can reduce dihydroflavonol, while the CsDFRb1 cannot. We used cDNA sequences from these

two genes to create five chimeric DFR genes in order to find the area that controls the activity

difference between CsDFRa and CsDFRb1. (Figure 6A and Dataset S2). Prior to catalytic analysis,

the purified proteins were prepared and purified. The reducing activity of DHQ was used as the

substrate to test the reducing activity of these chimeric DFR proteins (Figure 6B).

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



13

Figure 6. Truncation and reorganization that determines the putative activity switch region between CsDFRa and

CsDFRb1. (A) The model of amino acid position by truncation. FS1, FS2, FS3, FS4 and FS5 stands for five different

reorganization strategies, amino acid residues with red represent fusion between CsDFRa and CsDFRb1, substrate-binding

region and putative activity switch region are boxed in black and blue, respectively. (B) Analyze of the enzyme reaction with

CsDFRa, FS1, FS2, FS3, FS4, FS5 and CsDFRb1 in vitro and DHQ as the substrate.

The chimeric gene FS1 was constructed with 5’-CsDFRb1 and 3’-CsDFRa at the start of the

CsDFRa substrate-binding region, while FS2 was created with 58 amino acids before the

substrate-binding region. The results of activity tests revealed that cyanidin formation could be

observed in the reaction system with FS1 protein in vitro, but not in the reaction system with FS2

protein. We speculate that the DFR protein has an activity switch region prior to the

substrate-binding region. Based on this speculation, three chimeric genes were further constructed,

and the sequence of CsDFRb1 was gradually being replaced by CsDFRa before the

substrate-binding region. The activity test results showed that the FS3 and FS4 recombinant

proteins exhibited reductase activity, whereas the FS5 recombinant enzyme has lost its reductase

activity. Therefore, it was predicted that the amino acid residues between FS4 and FS5

recombination points might have significant impact on DFR enzyme activity.

Site-directed mutagenesis of the activity switch region

The amino acid sequence alignment demonstrated that the amino acids in the putative activity

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



14

switch (PAS) region of CsDFRa and CSDFRb1 are relatively conserved, with only four amino

acids difference (Figure 7A). We conducted a series of gene mutation studies on different residues

to further confirm that these distinct amino acids alter the DFR activity, and DHQ was used as a

substrate to evaluate the reducing activity of the mutated proteins (Figure 7B). These results

showed that the CsDFRa mutated proteins, CsDFRaS117N

and CsDFRaT123C

, lost the DHQ reducing

activity. Meanwhile, the DFR activity was restored by CsDFRb1 mutations, CsDFRb1N120S

and

CsDFRb1C126T

. The mutations in the other two different residues, such as DFRaI118V

and

DFRaI119M

, had no effect on the activity of the non-mutated protein (data no shown). The results of

site-directed mutagenesis showed that the serine at 117 and threonine at 123 of CsDFRa are

critical for modulating DFR reducing activity. Interestingly, these key amino acids are both belong

to the hydroxyl amino acid family.

CsDFRa and CsDFRb1 homologous modeling was carried out using dihydroquercetin (DHQ) and

NADPH as substrates, and splicing experiments were conducted using grape DFR crystals (PDB

code 3BXX) 26

. There was no significant difference in the binding of DFR residues to the substrate

and NADPH in CsDFRs and site-directed mutated proteins (Figure 7B). Furthermore, the PAS

region was found in helix α4, and the X-ray 3D structure revealed that this helix promotes water

molecule binding to the carbonyl group of the asparagine main chain and the amino group of the

catalytic lysine 27

. This is the first of a cluster of water molecules that transfers a proton from the

bulk solvent to the catalytic lysine 28,29

.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



15

Figure 7. Identification of key amino acids that is critical for modulating DFR reducing activity. (A) Analysis of enzymes

activity with different site-directed mutagenesis of the activity switch region. The red asterisk indicates the different amino acids

of DFRa and DFRb1 at the putative activity switch region, DHQ was used as a substrate to evaluate the reducing activity of the

mutated proteins. (B) Homology model of DFRa, DFRb1 and site-directed mutagenesis with docked dihydroquercetin (DHQ)

and NADPH ligand, the PAS region was located in the helix α4 (red band). The model of homology was built using grape DFR

crystals as a template (PDB code 3BXX).

Catalytic activities of recombinant CsDFR proteins in vitro

The dihydroflavonol substrate kinetic parameters of recombinant CsDFR proteins, including

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



16

CsDFRa, CsDFRc, CsDFRb1N120S

and CsDFRb1C1267T

, were determined in potassium phosphate

buffer at pH 7. CsDFRa Km values with DHK, DHQ, and DHM were 145.10, 41.80, and 58.44

μM, respectively. Meanwhile, for CsDFRc, the Km values for DHK, DHQ, and DHM were 42.31,

81.80, and 105.56 μM, respectively. The results indicated that CsDFRa may have a higher affinity

for DHQ, whereas CsDFRc has a higher affinity for DHK (Figure 8 and Table S3).

Enzymatic assays were used to evaluate the activities of the site-directed mutated proteins of

CsDFRb1N120S

and CsDFRb1C126T

. CsDFRb1N120S

Km values for DHK, DHQ, and DHM were

148.96, 108.66, and 96.58 μM, respectively, while CsDFRb1C126T

Km values were 131.09, 102.57,

and 47.75 μM. These findings revealed that the N120S and C126T substitution for CsDFRb1

exhibit a higher affinity for the substrate DHM (Figure 8 and Table S3).

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



17

Figure 8. Kinetic parameters for DFRa, DFRc, DFRb1N120S and DFRb1C126T. In each enzyme reaction assay Two two

micrograms with purified recombinant proteins were used.. A uniform concentration of 0.05 mg dihydroflavonol (DHK, DHQ

and DHM) was used as substrate. The data are expressed as means ± SD (n=3).

Discussion

CsDFR involvement in the biosynthesis of flavonoids in C. sinensis

DFR catalyzes the NADPH-dependent reduction of three different dihydroflavones to form the

corresponding leucoanthocyanins 10

. Leucoanthocyanins are the precursors for the synthesis of

anthocyanins and flavan-3-ols. Therefore, DFR is the pivotal enzyme responsible for anthocyanins,

catechins, and proanthocyanidins accumulation in plants. Catechins, procyanidins, and flavonols

are the most abundant flavonoids in C. sinensis, with catechins mainly accumulated in buds and

young leaves, while procyanidins are mainly accumulated in roots 7,30

.

Numerous studies suggest that DFR expression is positively correlated with anthocyanin and PAs

accumulation. In Saussurea medusa, for example, the SmDFR gene was expressed highly in

flowers but significantly low in leaves and roots 31

. In Lilium asiatica, two DFR genes were highly

expressed in the colored parts of the perianth, anther, filament, stigma, and scale 32

. Shimada

investigated the tissue expression specificity of DFR using Bermuda fruit, flower, leaf, stem, and

root as materials, and demonstrated that the tissue expression specificity of DFR gene was

consistent with its catalytic activity 33

. Tang LK demonstrated that AtDFR was highly expressed

during pollen tube development. The results suggested that AtDFR may play an important role in

pollen maturation and growth 34

. In tea plants, Mamati showed that the DFR gene was often highly

expressed in tissues with high catechin content 35

.

Four DFR genes were identified and described from C. sinensis in this work. Among those genes,

CsDFRa is strongly expressed in the bud and young leaf of C. sinensis, implying that CsDFRa is

the major gene determining the accumulation of catechins in tea plant. While the other genes, on

the other hand, are significantly expressed in the young root of tea plant, indicating that the three

genes are linked to PAs buildup in the root.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



file:///D:/QQé�³ä¹�/Youdao/Dict/7.5.2.0/resultui/dict/?keyword=enzyme

file:///D:/QQé�³ä¹�/Youdao/Dict/7.5.2.0/resultui/dict/?keyword=enzyme

file:///D:/QQé�³ä¹�/Youdao/Dict/7.5.2.0/resultui/dict/?keyword=reaction

18

Catalytic specificity analysis of the DFR family

Dihydroflavonol is classified into three types based on the number hydroxyls groups on the B-ring,

they include DHK (mono-hydroxyl in B-ring), DHQ (di-hydroxyl in B-ring) and DHM

(tri-hydroxyl in B-ring). In this study, enzyme activity analysis showed that four DFR enzyme

exhibited different selectivities to three substrates in vitro. According to the structure analysis of

proteins, DFR contains two important domains: the N-terminal highly conserved “NADPH”

binding site and the substrate “dihydroflavones” binding site 36

.

Previous studies has shown that the amino acid 134 is important for the catalytic activity of DFR

protein 15

. Therefore, based on the type of amino acid 134, DFR can be divided into three

categories. Type 1, when position 134 is Aspartic acid (Asp), it is called the Asp-type, which can

catalyze DHK to form the leucoanthocyanins. This type of DFR protein has been reported in

cranberry 37

. Type 2, when the position of amino acid 134 is asparagine (Asn), it is called the

Asn-type, which can't catalyze the DHK to form leucoanthocyanins. Type 3, when position of

amino acid 134 is neither Asp nor Asn, it is called non-Asp or Asn-type.

With the development of bioinformatics, DFRs have been identified in numerous plant species.

Meanwhile, the previously identified genes of DFR family can also be divided into different types.

For instance, in the Lotus corniculatus Linn, five DFR genes can be divided into three types.

Among them, LcDFR1 belongs to the non-Asp or Asn-type, LcDFR2 and LcDFR3 belong to the

Asp-type, the other two, LcDFR4 and LcDFR5, belong to the Asn-type 38

.

In this study, we found that the understanding of the basic profile of DFR involvement in the

catechin biosynthesis in C. sinensis is still being unraveled. The evolutionary analysis shows that

CsDFRa belongs to Class I, and the function of most genes in this subfamily has been confirmed

to have dihydroflavonol reductase activity (Figure 2B). Members of Class I are classified as Asn

or Asp types based on the difference in the key amino acid sites in the substrate binding region.

The kinetic analysis of recombinant enzyme revealed that CsDFRa had a high DHQ conversion

efficiency.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



19

The rooted phylogenetic tree analysis shows that CsDFRb1, CsDFRb3 are grouped into the Class

II and CsDFRc are grouped into class III, which belongs to the non-Asp or Asn-type. The

enzymatic activities of members of this subfamily have not been reported before. In the present

study, we demonstrate that CsDFRc has the catalytic activity of dihydroflavonol 4-reductase by

the functional analysis of prokaryotic recombinant protein and Arabidopsis transgenic in vitro and

in vivo, respectively. The kinetic analysis of the recombinant enzyme showed that CsDFRc

displayed highest conversion efficiency for DHK. However, CsDFRb1 and CsDFRb3 did not

detect the catalytic activity of Dihydroflavonol 4-reductase.

Crucial amino acid residues or regions for DFR protein

Several studies have indicated that the DFR substrate specificity is determined by a 26 amino acid

region 15

. CsDFRb1 without catalytic activity and CsDFRa with catalytic activity were truncated

and recombined for each other, and the recombinant protein was expressed in E. coli to further

investigate the key amino acid residues that determine the function of dihydroflavonol reductase.

Analysis of enzyme activity showed that the twenty amino acid residues before the substrate

binding region might have important effects on DFR enzyme activity. Furthermore, site-directed

mutation confirmed that N120 and T126 are the key amino acid sites determining the catalytic

activity of CsDFRb. Interestingly, among them, CsDFRb1N120S

and CsDFRb1T126C

have catalytic

activity, and their affinity for DHM is significantly higher than that of DHQ and DHK. These

results can be used for metabolic engineering of B-ring trihydroxyflavonoid.

Homologous modeling and substrate docking showed that the site-directed mutation had no effect

on the three-dimensional structure of the DFR protein, and there was no significant difference in

the spatial conformation between the substrate and NADPH. The putative activity switch (PAS)

region is located in helix α4 region (Figure 7B), which mainly determines the second H+ supply in

the reduction reaction 27

. Our findings confirmed that the key amino acid residues, serine and

cysteine in PAS region (S117 and C123 of CsDFRa), are mercapto amino acid or hydroxyl amino

acid, which may participate in the transfer of H+. The hydroxyl group established H-bond with

water molecule or protein is critical for maintaining the reductase activity. The establishment of

the H-bond with water molecule and proteins can’t be refuted, because the binding of a water

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



20

molecule to the carbonyl group of the main-chain of the asparagine (N120 of CsDFRb1) and the

amino group of the catalytic lysine (T126 of CsDFRb1), they either break H-bond or induce steric

hindrance for substrate binding, which coincides with previous findings 15

. Taken together, these

studies provide a theoretical foundation for learing more about DFR’s catalytic.

Conclusion

In conclusion, we isolated and characterized the CsDFR genes (CsDFRa, CsDFRc, CsDFRb1 and

CsDFRb3) in C. sinensis and demonstrated their roles in vitro and in vivo. We found that its

function is closely associated with catechins and PAs accumulation. Its physiological role was

studied in the Arabidopsis tt3 mutant, and it was found that CsDFRa and CsDFRc recovered tt3

mutant phenotypes. Moreover, the enzyme activity assay was determined in vitro for the DFR

activity of the CsDFRs and found that CsDFRa and CsDFRc have the DFR activity which

converts dihydroflavonols into leucocyanidin. Through site-directed mutations, it was found that

two amino acid residues of CsDFRs play an important role in enzyme activity. The enzymes

kinetic revealed that the recombinant CsDFR enzyme showed that the optimal substrates for

CsDFRa and CsDFRc were DHQ and DHK, and the optimal substrates for CsDFRb1N120S

and

CsDFRb1C126T

were DHM, respectively. Thus, our data suggests that DFRs evolutionarily

conserved in many plant species, which helps to provide a new candidate gene for the metabolic

engineering of the flavonoids.

Methods

Plant materials

Different tissues bud, young leaf, mature leaf, stem and root of C. sinensis cv. Shuchazao were

collected in early autumn from a five years old plant at Anhui Agricultural University in Hefei,

China (north latitude 31.86, east longitude 117.27, altitude 20 m above mean sea level)

experimental garden. The samples were frozen and stored at -80°C for future use.

The Landsberg erecta (Ler, wild-type) and AtDFR mutant (tt3, CS84) of Ler Arabidopsis were

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



21

obtained from the Arabidopsis Biological Resource Center (abrc.osu.edu) and grown in the growth

chamber of the College of Life Sciences, Anhui Agricultural University with a 16 h light/8 h dark

cycle at 22±2°C with a light intensity of 200 μmol m-2

s-1

.

Isolation and cloning of CsDFR genes

RNAiso Plus for plant tissue kits (Takara, Dalian, China) was used to isolate total RNA from the

bud and root. Standard end-to-end PCR reaction were performed on CsDFRs genes from the

NCBI database, with gene specific primers designed (Table S2) according to the cDNA sequence.

End-to-end PCR was carried out at 98°C for 30 s, 32 cycles at 98°C for 10 s, 58°C for 15 s, 72°C

for 30 s, and a final extension at 72°C for 10 min. Subsequently, the PCR amplification products

were purified using a Gel Extraction Kit (Aidlab, Beijing, China) and cloned into an Easy-Blunt

vector (TransGen Biotech). The resulting vector was introduced into DH5α competent cells and

sequenced.

Phylogenetic analysis

The unabridged length amino acid sequences of DFR proteins from C. sinensis and several other

plants were acquired from the NCBI database (https://blast.ncbi.nlm.nih.gov/Blast.cgi) for

Phylogenetic analysis. ClustalW version 2.1 was used to align all of the sequences and deduced

amino acid sequences of CsDFRs, and the alignment was then submitted to MEGA7 software

(https://www.megasoftware.net) to generate a tree with neighbor-joining and bootstrapping (1000

replicates) analysis, as well as handling gaps with complete deletion.

Quantitative real-time PCR analysis

The gene-specific qRT-PCR primers were designed to study the tempo-spatial expression patterns

of CsDFR genes in C. sinensis (Table S2). On a Peltier Thermal Cycler PTC200 (Bio-Rad, USA),

the transcript level was determined using the SYBR premix Ex Taq II kit (Takara, Japan)

Triplicate of all biological samples were analyzed. The Ct values of the housekeeping gene

glyceraldehyde-3-phosphate dehydrogenase (GAPDH) from C. sinensis were used to normalize

the reaction.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



javascript:;

https://blast.ncbi.nlm.nih.gov/Blast.cgi

22

Heterologous expression of CsDFRs in A. thaliana

Using Gateway technology (Invitrogen, USA) the full-length CsDFRs cDNA was shuttled into the

pCB2004 binary vector. Subsequently, the positive recombinant pCB2004-CsDFRs vector was

Agrobacterium tumefaciens (GV3101) using electroporation, and the Agrobacterium was then

used for floral dip transformation of Arabidopsis mutant tt3 39

. The transformed plants were grown

in the glasshouse under carefully controlled conditions (16 h light/8 h dark cycle at 22±2°C with a

relative humidity of 40%).

Extraction and quantification analysis of anthocyanins and PAs from plant materials

For the extraction of anthocyanins and PAs, briefly, freeze-dried samples (including seedlings and

seeds) of 0.05 g of CsDFRs transgenic lines, WT and tt3 mutant were ground by grinders.

Subsequently, the total polyphenols were extracted at a low temperature with 1.2 mL of extraction

solution A (HCl 0.5%: methanol 80%: water 19.5%, v/v) to the samples and vortexed and

sonicated for 30 min at 4°C 7. The samples were centrifuged at 13000g for 15 minutes before

being re-extracted twice more as indicated above until the final volume of supernatants was 2mL.

Finally, the combined solution was centrifuged for 10 minutes at 13 000g and the supernatants

were stored at −20°C prior to total content analysis of anthocyanins and PAs. The contents of total

anthocyanins in the supernatants were measured with UV spectrometer at 530 nm. After a reaction

with the DMACA reagent (0.2% DMACA, w/v, HCl/methanol 1:7, v/v), the total soluble PAs

content in the seeds was determined by a UV spectrometer at 640 nm. Briefly, the reaction mixture

(1.3 mL total volume) contains 0.3 mL of supernatants and 1 mL of DMACA reagent.

Subsequently, the reaction mixture was incubated at 25°C for 5 min, and the absorbent values of

the reaction mixture were measured at 640 nm.

Heterologous expression of CsDFRs in Escherichia coli

The CsDFRs ORF was subcloned into the pET-SUMO vector (Invitrogen, California, USA) in its

entirety. The identity of the cloned gene was validated using T7 primers and sequencing analysis.

The resulting vector SUMO-CsDFRs were transformed into BL21 (DE3) E. coli competent cells,

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



23

according to the manufacturer’s instructions. All of the chimeric and point mutation genes were

cloned into the pET-SUMO vector with the same method as above. Positive colony, expression

strains were cultivated in 200 mL of LB medium with 50 μg·mL−1

kanamycin at 37°C. To

promote recombinant protein expression, 1 mM isopropyl-β-D-thiogalactoside (IPTG) was added

when the E.coil OD600 reached 0.6. The cells were collected by centrifugation after 8 hours of

induction at 28°C, and the fusion protein was purified using His-tag affinity chromatography

according to the manufacturer’s instructions. Twelve percent SDS polyacrylamide gel staining and

Coomassie brilliant blue staining were used to validate the isolated protein fraction. The activity

assay in vitro was performed using pure recombinant proteins. All specific primers for

SUMO-CsDFRs, chimeric DFRb1/DFRa, and amino acid point mutant genes were listed in Table

S2.

Enzyme activity assay and analysis of kinetics parameters

The enzymatic reaction consisted of 0.1M pH 7.0 potassium phosphate buffer, 2 mM NADPH, 0.5

mM dihydroflavonols (DHK, DHQ and DHM) and 0.05 mg total soluble proteins content. The

reaction mixtures were incubated at 30°C for 30 min, and the total volume was 200 μl. Because

leucoanthocyanidin is unstable in solution and difficult to detect directly using HPLC, a three

times volume of butanol–HCl reagent (95:5, v/v) was added, and the reaction was splitted at 95°C

for 1h to form anthocyanidin and was centrifuged at 4000g for 15min. The absorption of the

supernatants was measured at OD530 nm by UV spectrophotometer. According to the absorption

value of anthocyanin standard at different concentrations, the standard curve was established to

quantify the content of products in the sample. To determine the apparent Km value of the reaction

mixture of different substrates, 0.05 mg purified recombinant enzyme and three dihydroflavonols

at a series of concentrations. The enzymatic products were quantified as the standard curve of

anthocyanins standard. Apparent Km data were calculated using Hanes plots.

Molecular Docking statistic and site-directed mutagenesis

The homology model of CsDFRa and CsDFRb1 were constructed with the crystal structure of

grape DFR (PDB code 3BXX) as a template using the online server SWISS-MODEL

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



24

(http://swissmodel.expasy.org/interactive). Furthermore, dihydroquercetin and NADPH were

taken as the substrates, and they were docked into the model of CsDFRa and CsDFRb1 with the

lowest energy conformation, using the software AutodockTools-1.5.6

(http://mgltools.scripps.edu/downloads). The resulting spatial model structure of the CsDFRa and

CsDFRb1 protein, including binding sites and substrate linkage pockets, were analyzed and

visualized using Python 27 software. The site-directed mutation plasmids CsDFRaS117N

,

CsDFRaT123C

, CsDFRb1N120S

, CsDFRb1C127T

, DFRaI178V

and DFRaI179M

were generated using an

overlap extension PCR technology with CsDFRa and CsDFRb1 plasmids as the templates,

respectively. All of the specific oligonucleotide primers used to construct the mutations plasmids

is listed in Table S2.

Statistical methods

Three separate experiments or three biological replicates were used in each of the aforementioned

test. Using SPSS software, mean values were generated and then statistically analyzed using a

Student’s test. The two-tailed test of significance was used. Statistical significance was determined

by P-values of less than 0.05.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31770729,

31870676); the National Key R&D Program of China (2018YFD1000601); Collegiate

Collaborative Innovation Foundation of Anhui Province (GXXT-2020-081) and Joint Funds of the

National Natural Science Foundation of China (U21A20232).

Author contributions

RHX, SXX and WYS conceived and designed the research. WYS, GLP and XT conducted

experiments. SXX, RHX, LY, LT and DXL performed the experiments. RHX, RA and WYS

analyzed the data. RHX and DXL wrote the manuscript, RA and WYS edited and contributed to

the content.

Conflict of Interest

There are no conflicts of interest declared by any of the writers.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



http://swissmodel.expasy.org/interactive

25

Data availability statement

All data supporting the conclusions of this study may be found in the publication and its

supplemental materials, which are available online. Any additional relevant information can be

obtained from the corresponding author upon request (Yunsheng Wang).

Supporting information

Figure S1: Schematic representation of CsDFRs gene structures and promoter elements.

Table S1: Relative quantity of different flavonoid compounds in different tissues.

Table S2: List primers used in this study.

Table S3: Kinetic parameters for DFRa, DFRc, DFRb1N120S

and DFRb1C126T

.

References

1. Winkel-Shirley B. Flavonoid biosynthesis. A colorful model for genetics, biochemistry,

cell biology, and biotechnology. Plant Physiol. 2001;126:485-93.

2. Cabrera C, Artacho R, Gimenez R. Beneficial effects of green tea--a review. J Am Coll

Nutr. 2006;25:79-99.

3. Chen A Y, Chen Y C. A review of the dietary flavonoid, kaempferol on human health

and cancer chemoprevention. Food Chem. 2013;138:2099-107.

4. Dalgaard F, Bondonno N P, Murray K et al. Associations between habitual flavonoid

intake and hospital admissions for atherosclerotic cardiovascular disease: a

prospective cohort study. Lancet Planet Health. 2019;3:e450-e59.

5. Tanaka Y, Brugliera F. Flower colour and cytochromes P450. Philosophical

transactions of the Royal Society of London. Series B, Biological sciences.

2013;368:20120432.

6. Hua C, Linling L, Shuiyuan C et al. Molecular cloning and characterization of three

genes encoding dihydroflavonol-4-reductase from Ginkgo biloba in anthocyanin

biosynthetic pathway. PLoS One. 2013;8:e72017.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



26

7. Jiang X, Liu Y, Li W et al. Tissue-specific, development-dependent phenolic

compounds accumulation profile and gene expression pattern in tea plant [Camellia

sinensis]. PLoS One. 2013;8:e62315.

8. Bowers J E, Dangl G S, Vignani R et al. Isolation and characterization of new

polymorphic simple sequence repeat loci in grape (Vitis vinifera L.). Genome.

1996;39:628-33.

9. Helariutta Y, Elomaa P, Kotilainen M et al. Cloning of cDNA coding for

dihydroflavonol-4-reductase (DFR) and characterization of dfr expression in the

corollas of Gerbera hybrida var. Regina (Compositae). Plant Mol Biol.

1993;22:183-93.

10. Wang J, Wang Y, Wang S et al. Cloning and temporal-spatial expression analysis of

dfr gene from Scutellaria baicalensis with different colors. Sheng Wu Gong Cheng Xue

Bao. 2021;37:1312-23.

11. Wang L, Zhu Y, Wang P et al. Functional Characterization of a Dihydroflavanol

4-Reductase from the Fiber of Upland Cotton (Gossypium hirsutum). Molecules.

2016;21:32.

12. Bashandy H, Pietiainen M, Carvalho E et al. Anthocyanin biosynthesis in gerbera

cultivar 'Estelle' and its acyanic sport 'Ivory'. Planta. 2015;242:601-11.

13. Kubasek W L, Shirley B W, Mckillop A et al. Regulation of Flavonoid Biosynthetic

Genes in Germinating Arabidopsis Seedlings. Plant Cell. 1992;4:1229-36. ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



27

14. Johnson E T, Yi H, Shin B et al. Cymbidium hybrida dihydroflavonol 4-reductase does

not efficiently reduce dihydrokaempferol to produce orange pelargonidin-type

anthocyanins. Plant J. 1999;19:81-5.

15. Johnson E T, Ryu S, Yi H et al. Alteration of a single amino acid changes the substrate

specificity of dihydroflavonol 4-reductase. Plant J. 2001;25:325-33.

16. Alagawany M, Abd El-Hack M E, Saeed M et al. Nutritional applications and beneficial

health applications of green tea and l-theanine in some animal species: A review. J

Anim Physiol Anim Nutr (Berl). 2020;104:245-56.

17. Yao S S, Guo W F, Lu Y et al. Flavor characteristics of lapsang souchong and smoked

lapsang souchong, a special Chinese black tea with pine smoking process. J Agric

Food Chem. 2005;53:8688-93.

18. Asakawa T, Hamashima Y, Kan T. Chemical synthesis of tea polyphenols and related

compounds. Curr Pharm Des. 2013;19:6207-17.

19. Hu J, Zhou D, Chen Y. Preparation and antioxidant activity of green tea extract

enriched in epigallocatechin (EGC) and epigallocatechin gallate (EGCG). J Agric Food

Chem. 2009;57:1349-53.

20. Rani A, Singh K, Ahuja P S et al. Molecular regulation of catechins biosynthesis in tea

[Camellia sinensis (L.) O. Kuntze]. Gene. 2012;495:205-10.

21. Wang W, Zhou Y, Wu Y et al. Insight into Catechins Metabolic Pathways of Camellia

sinensis Based on Genome and Transcriptome Analysis. J Agric Food Chem.

2018;66:4281-93.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



28

22. Katsu K, Suzuki R, Tsuchiya W et al. A new buckwheat dihydroflavonol 4-reductase

(DFR), with a unique substrate binding structure, has altered substrate specificity.

Bmc Plant Biol. 2017;17:239.

23. Shirley B W, Hanley S, Goodman H M. Effects of ionizing radiation on a plant genome:

analysis of two Arabidopsis transparent testa mutations. Plant Cell. 1992;4:333-47.

24. Pang Y, Abeysinghe I S, He J et al. Functional characterization of proanthocyanidin

pathway enzymes from tea and their application for metabolic engineering. Plant

Physiol. 2013;161:1103-16.

25. Xie D Y, Sharma S B, Paiva N L et al. Role of anthocyanidin reductase, encoded by

BANYULS in plant flavonoid biosynthesis. Science. 2003;299:396-9.

26. Trabelsi N, Petit P, Manigand C et al. Structural evidence for the inhibition of grape

dihydroflavonol 4-reductase by flavonols. Acta Crystallogr D Biol Crystallogr.

2008;D64:883-91.

27. Petit P, Granier T, D'estaintot B L et al. Crystal structure of grape dihydroflavonol

4-reductase, a key enzyme in flavonoid biosynthesis. J Mol Biol. 2007;368:1345-57.

28. Benach J, Filling C, Oppermann U C et al. Structure of bacterial

3beta/17beta-hydroxysteroid dehydrogenase at 1.2 A resolution: a model for multiple

steroid recognition. Biochemistry. 2002;41:14659-68.

29. Bottoms C A, Smith P E, Tanner J J. A structurally conserved water molecule in

Rossmann dinucleotide-binding domains. Protein science : a publication of the Protein

Society. 2002;11:2125-37.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



29

30. Zhao J, Li P, Xia T et al. Exploring plant metabolic genomics: chemical diversity,

metabolic complexity in the biosynthesis and transport of specialized metabolites with

the tea plant as a model. Crit Rev Biotechnol. 2020;40:667-88.

31. Li H, Qiu J, Chen F et al. Molecular characterization and expression analysis of

dihydroflavonol 4-reductase (DFR) gene in Saussurea medusa. Mol Biol Rep.

2012;39:2991-9.

32. Nakatsuka A, Izumi Y, Yamagishi M. Spatial and temporal expression of chalcone

synthase and dihydroflavonol 4-reductase genes in the Asiatic hybrid lily. Plant

Science. 2003;165:759-67.

33. Shimada N, Sasaki R, Sato S et al. A comprehensive analysis of six dihydroflavonol

4-reductases encoded by a gene cluster of the Lotus japonicus genome. Journal Of

Experimental Botany. 2005;56:2573-85.

34. Tang L K, Chu H, Yip W K et al. An anther-specific dihydroflavonol 4-reductase-like

gene (DRL1) is essential for male fertility in Arabidopsis. New Phytol.

2009;181:576-87.

35. Mamati G E, Liang Y R, Lu J L. Expression of basic genes involved in tea polyphenol

synthesis in relation to accumulation of catechins and total tea polyphenols. J Sci

Food Agr. 2006;86:459-64.

36. Martens S, Teeri T, Forkmann G. Heterologous expression of dihydroflavonol

4-reductases from various plants. FEBS Lett. 2002;531:453-8. ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



30

37. Polashock J J, Griesbach R J, Sullivan R F et al. Cloning of a cDNA encoding the

cranberry dihydroflavonol-4-reductase (DFR) and expression in transgenic tobacco.

Plant Science. 2002;163:241-51.

38. Shimada N, Sasaki R, Sato S et al. A comprehensive analysis of six dihydroflavonol

4-reductases encoded by a gene cluster of the Lotus japonicus genome. J Exp Bot.

2005;56:2573-85.

39. Clough S J, Bent A F. Floral dip: a simplified method for Agrobacterium-mediated

transformation of Arabidopsis thaliana. Plant J. 1998;16:735-43.

ORIGIN

AL UNEDIT

ED MANUSC

RIPT

Dow



PDF - ORIGINAL UNEDITED MANUSCRIPT

Documents

Transcript of PDF - ORIGINAL UNEDITED MANUSCRIPT